the PEAR experiment debunked (again)

[Stimpson J. Cat]

jzs,

I presented a mathematical derivation

No. You speculated how a mathematical function would affect random data.

I did not speculate how a mathematical function would affect random data, I derived how the mathematical function in question does affect both random and correlated data, and by doing so have mathematically shown that the analysis method is flawed.

I've been asking you to demonstrate what you are claiming with actual data, specifically do it with the PEAR data.

What you are asking is nonsensical, because I have not been making claims about the data, but about the analysis method. As I have already explained to you repeatedly, even if the data is perfect, the analysis method being used is still flawed. The problem is the method, not the data.

And I could care less about your challenges, really. Either present actual evidence of your theory or don't.

I have not presented a theory. I am sorry that you do not understand that.

I can only conclude that you either do not understand my derivation, or you understand it, know that I am right, and do not want to admit it.

One thing I've noticed, when reading your posts in this thread and on the other board in the PEAR thread, is that you often tend to present things with two possibilities; where both possibilities are stated to make me look the worst. That's highly dishonest of you, and whenever you phrase something in a dishonest way like this in the future, it will be ignored.

As far as I can see, all you have been doing so far is ignoring me. Sure, you make posts which are phrased as responses to me, but your responses neither address the arguments I have made in any kind of meaningful way, nor do they even indicate that you have understood my arguments at all.

He is speculating that a 1010... mask is always used. That no shifts are used. That only a non-Mindsong mask is used. That there is a lot of correlation. That the bias is a severe problem. That we can't say anything useful from the binomial model. And on and on.

I have not made any such "speculation" or claims. Nor do any of my arguments require that any of the above be true. Again, I am not making claims about the data. My argument is about the analysis method, and what is wrong with it.


Dr. Stupid

 

[jj]

Re: Re: Re: a question

My point was that you can design a experimental protocol so that XOR is not needed (I think) a la the original PEAR studies. Unfortunately, the nature of the GCP means that it is littered with design problems that aren't anything to do with the output of the REG's. I don't know why they can't just perform controlled experiments instead.

Well, now that you put it that way, I agree 100%.

 

[T'ai Chi]

Sorry to just chime in, but it seems to me you're the one prolonging it, jzs. Skeptics point out flaws, and you never address them. One even handed out a mathematical proof of an inherent flaw in the data gathering. Show where his proof is wrong or admit that you were wrong. Unless you're trying to found your own branch of mathematics, it's not a "theory".

LOL, show me this "mathematical proof"..

His critique depends on the size of the bias. If it ain't large, his critique isn't worth much.

You're prolonging this whole thing with your silence.

That's a new one! Who knew saying nothing could be so damning!

 

[T'ai Chi]

Your continued insistance, your moving the goalposts, your failure to assume the burden you've taken up for yourself, and such are all terribly evident, as is your childish retaliation both here and in the other thread where you're stalking me, lying about context and content.

NOONECARESABOUTYOURSOAPOPERAS.

 

[T'ai Chi]

, I derived how the mathematical function in question does affect both random and correlated data,

What is the size of the correlation you are talking about, Stimpy?

 

[delphi_ote]

NOONECARESABOUTYOURSOAPOPERAS.

NOBODYCARESABOUTSPACEBARORCAPSLOCK

 

[davidsmith73]

Again, I am not making claims about the data. My argument is about the analysis method, and what is wrong with it.


Dr. Stupid

Your argument is indeed based on their analysis method but the validity of their analysis method is based on how the real data behaves. Therefore your argument does touch upon the actual data.

 

[Stimpson J. Cat]

jzs,

I derived how the mathematical function in question does affect both random and correlated data,

What is the size of the correlation you are talking about, Stimpy?

What correlation? I made a statement about what the XOR operation does to data, not about the degree of correlation in any specific data set.


davidsmith,

Again, I am not making claims about the data. My argument is about the analysis method, and what is wrong with it.

Your argument is indeed based on their analysis method but the validity of their analysis method is based on how the real data behaves. Therefore your argument does touch upon the actual data.

First of all, my argument about how the XOR operation does not do what they apperantly think it does, is not in any way dependent on any data.

As to the validity of their analysis, that does not depend on the data, it depends on what they have demonstrated to be true about the data.

In order for their analysis to be valid, it is not sufficient that the data meet the requirements of the statistical analysis methods they are using. It is necessary for them to demonstrate that it does. My argument does not depend on the data failing to meet the requirements of their analysis method. It depends on the fact that they have not demonstrated that it does meet those requirements. Indeed, they do not even seem to understand what those requirements are.

Like I said before, even if the biases in the real data are small enough that they do not significantly affect the results, and even if the anomalies found during some of these "global events" are due to some sort of influence of global consciousness, their analysis is still invalid, because they have not demonstrated that the bias is too small to significantly affect the results, nor have they shown that these anomalies are correlated with global events in any significant way.

Remember that even if there are no significant biases in the data produced by the REGs, short sequences of data which differ by a statistically significant amount from the average behavior of the REGs will happen from time to time, just due to chance. I have still yet to see any analysis showing that these anomalies occur more often than when be expected due to chance, nor have I seen any analysis showing that these anomalies are correlated with global events.


Dr. Stupid

 

[Bronze Dog]

NOONECARESABOUTYOURSOAPOPERAS.

Translation: I don't respond to legitimate criticism. I just keep shouting and repeating myself while contributing nothing of substance to the discussion.

 

[Zep]

Re: Re: Re: Re: a question

I do.

This way, they can obfuscate and still get funded, so they can continue.

Hey! That's MY line!

 

[davidsmith73]

First of all, my argument about how the XOR operation does not do what they apperantly think it does, is not in any way dependent on any data.

They apparently agree that the variance will be biased after XORing, which is what you have been saying isn't it? Lets not assume that they have misunderstood the results of XORing because they use the word "compensate" or "correct for" mean bias instead of "convert". The web site looks like it's for the general reader after all.

As to the validity of their analysis, that does not depend on the data, it depends on what they have demonstrated to be true about the data.

So, are you saying that it's possible that the data contains sufficiently little bias as to validate their analysis method, but they haven't shown that it does?

 

[jj]

They apparently agree that the variance will be biased after XORing, which is what you have been saying isn't it? Lets not assume that they have misunderstood the results of XORing because they use the word "compensate" or "correct for" mean bias instead of "convert". The web site looks like it's for the general reader after all.

Well, how do you propose, then, that this XOR'ing "compensate"s or "correct(s) for"?

What useful function does it perform. Can you be more specific?

If they are to expound results, they need to explain this, clearly and simply, and they haven't. The most we've heard is a second-hand dismissal reflected over at TOOP.

So, are you saying that it's possible that the data contains sufficiently little bias as to validate their analysis method, but they haven't shown that it does?

I can't see how they've shown (or for that matter not shown) anything. In my opinion there are rather a lot of problems with the PEAR design, this one alone is enough to dismiss it (although they could always start over with new data and a new method), and there are other reasons to dismiss it as well.

 

[Stimpson J. Cat]

davidsmith,

They apparently agree that the variance will be biased after XORing, which is what you have been saying isn't it?

Not exactly. They observe that the variance is still biased after XORing, but they do not give any indication that they realize that this is because XORing mean biased data results in a correlated bitstream. I would certainly hope that if they did realize this, they would at least address the fact that the XORed data is not uncorrelated. I would also hope that they would realize that simply normalizing to the estimated variance is not sufficient to make sure that these correlations do not bias the probability estimates.

In addition, they say this:

After XOR'ing, the mean is guaranteed over the long run to fit theoretical expectation. The trial variances remain biased, however. The biases are small (about 1 part in 10,000) and generally stable on long timescales. They are corrected by standardization of the trialsums to standard normal variables (z-scores). Mindsong regs tend to have positive biases. This gives a net positive variance bias to the data. Since the GCP hypothesis explicitly looks for a positive variance bias, there is a requirement for important, albeit small, corrections.

If they recognized that these biases in the variance were due to correlations in the data, then surely they would realize that their normalization will only eliminate any net positive or negative bias in the variance. It will not correct any abnormalities in the variability of the variance, which will also directly affect their analysis. They should at least have addressed this.

Also consider this:

The variance biases tell us that the raw, unXOR'd trials cannot be modeled as a simple binomial with shifted bit probability.

This statement pretty clearly indicates that they do not realize that XORing the data will introduce correlations, because the fact is that a simple binomial with shifted bit probabilities will, after XORing, have a bias in the variance. They appear to believe that the bias in the variance is a completely separate issue to the bias in the mean.

Lets not assume that they have misunderstood the results of XORing because they use the word "compensate" or "correct for" mean bias instead of "convert". The web site looks like it's for the general reader after all.

Well, I agree that it is possible that they do understand what the effects of XORing a bit stream are, and that I have misinterpreted what they have said about it, but if this is the case, then their attempts to normalize the XORed data make absolutely no sense. If they know that the XORed data is not an uncorrelated binomial process, then it makes absolutely no sense to compare it against the model of an uncorrelated binomial process. The only rationalization I can think of for this would be if they think that simply normalizing so that the variance matches that of an uncorrelated binomial process, somehow makes sure that all of its other statistical properties will match such a process too.

Either way, their overall normalization procedure does not do what they think it does. It does not turn an uncorrelated bitstream with a biased mean into an uncorrelated binomial process.

As to the validity of their analysis, that does not depend on the data, it depends on what they have demonstrated to be true about the data.

So, are you saying that it's possible that the data contains sufficiently little bias as to validate their analysis method, but they haven't shown that it does?

Close, but not exactly.

My argument concerning the validity of their analysis does not require that their data not meet the requirements of the analysis, only that they have not demonstrated that it does. That does not mean that I do not also think that the data does not meet the requirements. I don't.

We already know that there are anomalies in the data, due to all sorts of effects. They even attempt to remove these anomalies from the control data they use to estimate their normalization factor. So there is really no question that the data does not meet the requirements of their analysis method. But that is not the point of my argument.

Imagine for a moment that they have complete knowledge of all of the potential "mundane" sources of anomalies, and that they have reliable method for identifying when such anomalies occur. They could then make sure that any anomalies found in the experimental data are not due to such mundane effects, and could also make sure that no such anomalies contaminate the control data.

If this were the case, then we could imagine that the control data, with these anomalies removed, might meet the requirements of their analysis. That is, the bias in the data from the REGs might be small enough so as to not have any significant affect on their results.

But even then, they would still need to demonstrate that it does. They would still need to show that the remaining biases are too small to affect their results. And they have not done this.

But since they cannot properly control for mundane anomalies in the output of the REGs, the whole point is really moot. This is really the fundamental problem with the entire idea behind this research. They know that, for reasons which are completely explainable using known physics, the REGs will occasionally produce streams of anomalous data. Their entire methodology is based on the fundamentally flawed notion that they can somehow control for all possible normal sources of such anomalies, and therefore determine that any remaining anomalies are paranormal.

This is, unfortunately, a common theme in parapsychology. The basic idea is that if they can rule out all normal explanations, the real explanation must be paranormal. The very strategy is flawed from the beginning. You can't rule out all possible natural explanations. You can't control for all possible external influences on an experiment.

Think about it. Even if there were no problems at all with their normalization process, and even if they could show that the statistical behavior of the REGs in the absence of external influences perfectly matches their model, what could these experiments possibly tell us? The only thing a rejection of the null-hypothesis would tell us is that an external influence was present. But so what? We already know that the REGs can be influenced by external things. How can you rule out all possible external influences? There is no way, even in principle to do this. Even if you ruled out every single external influence you can think of, what about all the ones you didn't think of?

The entire concept of this research is simply nonsense. There is absolutely no reason to imagine that there is some global consciousness out there. There is absolutely no reason to imagine that it might influence the behavior of Random Number Generators. And there is absolutely no way to reliably test the hypothesis that it is, without being able to control for all other possible external influences, which is impossible. The entire project was doomed to failure before it even started.


Dr. Stupid

 

[davidsmith73]

Stimpson J Cat,

Thanks for your replies to my questions. I have one more. Would most of the problems we've been discussing regarding XORed correlations and non-stationarity be resolved if they performed a correlation analysis on "global world events" and their statistical anomalies, and using empirical control data, like you suggested?

 

[Stimpson J. Cat]

davidsmith,

Thanks for your replies to my questions. I have one more. Would most of the problems we've been discussing regarding XORed correlations and non-stationarity be resolved if they performed a correlation analysis on "global world events" and their statistical anomalies, and using empirical control data, like you suggested?

Well, it would resolve some of them, in that they would no longer be relevant to the analysis at all.

Let me put it this way. If I, for some reason, wanted to determine whether or not there was some sort of correlation between global events, and the occurrences of anomalous behavior in these devices, I would do something like the following:

First, I would come up with some sort of consistent, clear, and unambiguous definition of what constitutes a "global event". This would be quite a challenge in itself, but it is very important that it be clear and objective. There must be no subjectivity involved in deciding whether an event qualifies, nor in deciding when it begins and ends.

Next, I would record data from the devices over a long period of time, while also recording the occurrence of global events.

Then, after collecting a large amount of data, I would perform a non-parametric correlation analysis between the two sets of data. Actually, I would perform many such tests, including things like cross-correlation and mutual information measures.

This way there would be no need to worry about biases in the REGs, or about anomalies in the data, or about how well the data fits some model. The only question being considered is whether or not there is some correlation between the behavior of the REGs, and the timing of these global events.

Of course, positive results from such a test would not say anything at all about global consciousness, or any other paranormal hypotheses. It would simply mean that some external influence which is correlated in some way with these events, is somehow affecting the REGs. After determining that such correlations exist, we could worry about trying to figure out what causes them.

Anyway, that's how I would do it, if I actually thought there was any point in wasting my time doing so.

Let's face it. If there is some strange currently unknown way in which consciousness influences the seemingly random behavior of quantum mechanical systems, that fact is not going to be discovered by people searching desperately for anomalies through data which is known to produce anomalies for perfectly understood reasons. It is going to be discovered by particle physicists, who are doing extremely careful and precise experiments on systems which are known to conform to the theoretical models with extraordinary accuracy.

This is the problem I have with parapsychology in general. I have no doubt that there are problems with our current scientific models, some of which will probably lead to important breakthroughs which will completely change the way we view the natural world. But these discoveries are going to be made by physicists, biologists, and chemists, by continuously pushing against the boundaries of what we know, and continuously testing their theories under more and more extreme conditions. They are not going to be made by searching for statistical anomalies in random data sets, nor are they going to be made by watching people perform conjuring and mentalist tricks in the hopes of finally finding somebody who can "do it for real". The parapsychologists are going about things in completely the wrong way. They are not trying to learn new things about the world around us. Instead, they are desperately searching for support for the mythical beliefs which they already hold.


Dr. Stupid

 

[davidsmith73]

davidsmith,


Well, it would resolve some of them, in that they would no longer be relevant to the analysis at all.

Let me put it this way. If I, for some reason, wanted to determine whether or not there was some sort of correlation between global events, and the occurrences of anomalous behavior in these devices, I would do something like the following:

First, I would come up with some sort of consistent, clear, and unambiguous definition of what constitutes a "global event". This would be quite a challenge in itself, but it is very important that it be clear and objective. There must be no subjectivity involved in deciding whether an event qualifies, nor in deciding when it begins and ends.

Next, I would record data from the devices over a long period of time, while also recording the occurrence of global events.

Then, after collecting a large amount of data, I would perform a non-parametric correlation analysis between the two sets of data. Actually, I would perform many such tests, including things like cross-correlation and mutual information measures.

This way there would be no need to worry about biases in the REGs, or about anomalies in the data, or about how well the data fits some model. The only question being considered is whether or not there is some correlation between the behavior of the REGs, and the timing of these global events.

Of course, positive results from such a test would not say anything at all about global consciousness, or any other paranormal hypotheses. It would simply mean that some external influence which is correlated in some way with these events, is somehow affecting the REGs. After determining that such correlations exist, we could worry about trying to figure out what causes them.

Excellent suggestions. Any method of analysis whereby internal biases of the REG's are not important is a step in the right direction. Statistical analysis aside, the main reservation I have about the GCP is the subjective nature of what constitutes a "global event". However, they seem to be specifying the time period of such events in advance of the data analysis so claims of data selection are more difficult to hold up.

Anyway, that's how I would do it, if I actually thought there was any point in wasting my time doing so.

Let's face it. If there is some strange currently unknown way in which consciousness influences the seemingly random behavior of quantum mechanical systems, that fact is not going to be discovered by people searching desperately for anomalies through data which is known to produce anomalies for perfectly understood reasons. It is going to be discovered by particle physicists, who are doing extremely careful and precise experiments on systems which are known to conform to the theoretical models with extraordinary accuracy.

This is where I disagree. Particle physicists may find such effects but they are less likely to because they are not specifically looking for such effects. They will most probably miss them (assuming they are there) or attribute them to other sources than consciousness. No, its the people who are actively interested in the possibility of the influence of consciousness on physical systems who are more likely to discover a reliable and replicable effect.

This is the problem I have with parapsychology in general. I have no doubt that there are problems with our current scientific models, some of which will probably lead to important breakthroughs which will completely change the way we view the natural world. But these discoveries are going to be made by physicists, biologists, and chemists, by continuously pushing against the boundaries of what we know, and continuously testing their theories under more and more extreme conditions. They are not going to be made by searching for statistical anomalies in random data sets, nor are they going to be made by watching people perform conjuring and mentalist tricks in the hopes of finally finding somebody who can "do it for real". The parapsychologists are going about things in completely the wrong way. They are not trying to learn new things about the world around us. Instead, they are desperately searching for support for the mythical beliefs which they already hold.

Well, not really. They are trying to form tentative hypotheses for phenomena that may have something to do with experience, consciousness and the "transfer" of information into these things. I have touched upon thishere

 

[Stimpson J. Cat]

davidsmith

Let's face it. If there is some strange currently unknown way in which consciousness influences the seemingly random behavior of quantum mechanical systems, that fact is not going to be discovered by people searching desperately for anomalies through data which is known to produce anomalies for perfectly understood reasons. It is going to be discovered by particle physicists, who are doing extremely careful and precise experiments on systems which are known to conform to the theoretical models with extraordinary accuracy.

This is where I disagree. Particle physicists may find such effects but they are less likely to because they are not specifically looking for such effects. They will most probably miss them (assuming they are there) or attribute them to other sources than consciousness. No, its the people who are actively interested in the possibility of the influence of consciousness on physical systems who are more likely to discover a reliable and replicable effect.

See, I think the situation is just the opposite. I think that it is precisely because the parapsychologists are looking for something so specific that creates a problem. You worry that the physicists may miss the effects, because they are not looking for them specifically, but the fact is that the physicists are looking for anything at all that they can find. They are searching for any new information which may be there, and performing any experiments they can think of which may yield new information. It is the parapsychologists who are likely to miss any real effects which may be there, both because they are using nearly hopelessly flawed methodologies for finding them, but also because they already have preconceived notions of what they are looking for. They are looking for something specific, so if what is really there does not match up with what they are looking for, they are almost certain to completely miss it.

Any deviation from the standard model of QM would be surprising to a physicist. And they really want to be surprised. The parapsychologists already believe that certain paranormal phenomena exist, and are looking for evidence to support those specific beliefs. The physicists have no idea what to expect, and will be thrilled to find anything at all. And rather than trying to message their data to match up with their pet psi theory, they will attempt to systematically determine what the new phenomenon is, and how it works.

Well, not really. They are trying to form tentative hypotheses for phenomena that may have something to do with experience, consciousness and the "transfer" of information into these things. I have touched upon thishere

They construct tentative hypotheses for phenomena which have not even been determined to actually exist. The first step is finding some phenomena which our current scientific theories cannot account for. Then you can start trying to find new theories to explain them.

Like I said before, the problem is not that parapsychologists are searching for phenomena which have no current scientific explanation. Physicists do this too. The problem is how they go about trying to do it. It is the methodology which is flawed. And if you think about it, methodology is all that really separates parapsychology from other sciences. For example, what the PEAR people are doing is parapsychology, but the experiment I suggested would be physics. What is the difference? We are both looking for a link between global events and the behavior of REGs. The difference is simply the methodology.

In principle, I would say that there is no such science as parapsychology. Any of the research being done which is being done properly, using proper scientific methodologies, will fit into one or more of the standard scientific fields, such as physics, chemistry, biology, psychology, or astronomy. There is no need for a separate field called "parapsychology". Indeed it seems as though the separate category only exists for purposes of image and funding. For example, this PEAR nonsense would never stand up to scrutiny in a physics journal. But technically speaking what they are doing, or at least trying to do, is physics. If there is an actual effect their, it would certainly be worthy of publication in a physics journal, but what they have would never get past the pear review process. At least, I certainly hope it would not. Certainly it would not be accepted by the physics community.


Dr. Stupid

 

[davidsmith73]

Originally posted by Stimpson J. Cat

See, I think the situation is just the opposite. I think that it is precisely because the parapsychologists are looking for something so specific that creates a problem. You worry that the physicists may miss the effects, because they are not looking for them specifically, but the fact is that the physicists are looking for anything at all that they can find. They are searching for any new information which may be there, and performing any experiments they can think of which may yield new information. It is the parapsychologists who are likely to miss any real effects which may be there, both because they are using nearly hopelessly flawed methodologies for finding them, but also because they already have preconceived notions of what they are looking for. They are looking for something specific, so if what is really there does not match up with what they are looking for, they are almost certain to completely miss it.

I appreciate that physicists are looking for anything that they can find, but I think its fair to say that "anything" is limited by the expectations of the individual and by what context they are working in. When a physicist searches for new information, they do so within the limits of a particular theoretical framework or at least they have a theoretical framework in mind to compare any unexpected results with. This means that they also have preconceived notions of the type of observations they are looking for. The trouble with searching for mind/matter effects is that there is no theoretical framework that includes a fundamental role of consciousness to compare to. This means that the researcher must start almost from scratch and base the initial stages of investigation on extremely tentative and basic hypotheses. Yes, they have preconceived notions of what they are looking for but all experimenters must have this to a degree.

Any deviation from the standard model of QM would be surprising to a physicist. And they really want to be surprised. The parapsychologists already believe that certain paranormal phenomena exist, and are looking for evidence to support those specific beliefs. The physicists have no idea what to expect, and will be thrilled to find anything at all. And rather than trying to message their data to match up with their pet psi theory, they will attempt to systematically determine what the new phenomenon is, and how it works.

Firstly, I am not going to make a generalisation about parapsychologists massaging data, because I don't see evidence for such a generalisation. Sure, there are cases of this happening but is it the case most of the time? Secondly, I think that most parapsychologists do not believe that these phenomena exist. They are trying to find out if these phenomena exist. In order to do this they must set up experiments so that if they exist they will detect such phenomena . That's an important distinction. Thirdly, I don't think its true that physicists have "no idea what to expect" when they perform exploratory experiments. Any type of experiment demands at least some notion of what is being observed. Sure, the precise nature of the results may not be expected but the experiment must be structured around a minimum level of predicted observation, otherwise they don't have an experiment!

They construct tentative hypotheses for phenomena which have not even been determined to actually exist. The first step is finding some phenomena which our current scientific theories cannot account for. Then you can start trying to find new theories to explain them.

Yes, the experiments are constructed to test for the existence of certain phenomena. Of course they have not been determined to exist already, otherwise such experiments would not be needed. However, the existence of such phenomena are suggested from reported experiences. At least, they are suggested enough to excite a certain proportion of scientists into performing controlled experiments.

Like I said before, the problem is not that parapsychologists are searching for phenomena which have no current scientific explanation. Physicists do this too. The problem is how they go about trying to do it. It is the methodology which is flawed. And if you think about it, methodology is all that really separates parapsychology from other sciences. For example, what the PEAR people are doing is parapsychology, but the experiment I suggested would be physics. What is the difference? We are both looking for a link between global events and the behavior of REGs. The difference is simply the methodology.

The methodological difference that separates parapsychology from other sciences is that parapsychology takes into account new hypothesised roles of consciousness into the experimental design. That's what should separate this field if we want to play at scientific categorisation. Flawed methodology does happen in parapsychology but is not what separates this field from others. And yes, what PEAR are doing is parapsychology because they are hypothesising a role of consciousness in the observed effect whereas your experiment is not.

In principle, I would say that there is no such science as parapsychology. Any of the research being done which is being done properly, using proper scientific methodologies, will fit into one or more of the standard scientific fields, such as physics, chemistry, biology, psychology, or astronomy. There is no need for a separate field called "parapsychology". Indeed it seems as though the separate category only exists for purposes of image and funding. For example, this PEAR nonsense would never stand up to scrutiny in a physics journal. But technically speaking what they are doing, or at least trying to do, is physics. If there is an actual effect their, it would certainly be worthy of publication in a physics journal, but what they have would never get past the pear review process. At least, I certainly hope it would not. Certainly it would not be accepted by the physics community.

I don't care much for games of categorisation. Is it physics/is it chemistry, is it neuroscience/is it psychology. At the end of the day, it bears nothing on the experiment. I think it is relavent when funding is taken into consideration. I think your view that parapsychology exists as a field for the purposes of funding could be true, it could not. If it is then I don't see a problem in that anyway. It's not like other fields don't compete for their piece of the pie!

 

[Garrette]

I am not even remotely in the same league as either Stimpson J. Cat or davidsmith73 when it comes to discussing the technical aspects of all this, but I feel compelled to address a couple of davidsmith's comments.

Originally posted by davidsmith73:

The trouble with searching for mind/matter effects is that...

I think this is the crux of it.

You have parapsychologists looking for "mind/matter effects."

Stimpson J. Cat emphasizes (and I agree) looking for "effects."

Once the effects are established to actually exist, then you can explore hypotheses such as mind/matter.

You have decided going into the experiment that any effects are mind/matter.

Originally posted by davidsmith73:

I don't care much for games of categorisation. Is it physics/is it chemistry, is it neuroscience/is it psychology. At the end of the day, it bears nothing on the experiment.

Obviously, though, you do care for categorization.

You have determined that physicists are ill-suited for this type of experiment.

 

[davidsmith73]

Once the effects are established to actually exist, then you can explore hypotheses such as mind/matter.

If mind/matter effects have been established to exist then this must have been done with a hypothesis that invloves mind/matter effects!

Unless you're talking about some other effect. Which would be what?

You have decided going into the experiment that any effects are mind/matter.

Not really. It should all be data collection with a very tentative hypothesis. The possibility that the observed effects are not due to mind/matter effects should not be ignored. If the data supports the tentative hypothesis then refine the hypothesis and test it thoroughly, as well as alternatives.

Obviously, though, you do care for categorization.

You have determined that physicists are ill-suited for this type of experiment.

I said I don't care much! I don't care for where the boundaries lie. That's a subjective argument.